Takuma Seno
Sep 5, 2018 · 1 min read

Hi, thank you for reading my article!

DQN has the loss function that is exactly same form as classical Q learning, which calculates TD errors.

Both of DQN and the classical Q learning use predicted values at the next state and the immediate reward. So, basically both do the same thing.

    Takuma Seno

    Written by

    A graduated computer science student at Keio University. My research theme is deep reinforcement learning. I’m working at SONY as a machine learning internship.

    Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
    Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
    Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade